Comment 28 for bug 1704396

Thanks, Michele.

One interesting thing from this log snippet is that there's no "flushing data" message -- though, I suspect that's simply because it was a C server (cstore) that stalled, but only Perl servers have that log message.

For the same reason (C server rather than Perl server) you wouldn't see the new log message. If you run across another instance where the search itself (the "Firing off the multiclass query" / "Returned from the multiclass query" pair) is where the long pause happens, that would definitely be a nearly-completely-Perl-stack API call, so would be a good candidate for spotting the new error message.

Another interesting thing is that the search service started data from the loading the ils_events file, meaning that it was trying to react to some error condition -- but what that was, I don't think we can tell.

I don't have any new theories based on this data, yet.